Our vision is to design the first truly mobile molecular brain imager that can be used on healthy subjects to study the functioning of the human brain during motion. The ultimate goal is to be able to image subjects during a proverbial walk in the park and other natural activities. We selected PET technology as the most likely to succeed in the next decade to provide the desired functionality of such a brain imager. While MRI is an exceptionally powerful and versatile imaging modality, and there are even upright MRIs for structural brain scans, for functional fMRI scans the subjects must stay still and in horizontal position inside a narrow bore of a strong-field MRI magnet. What we propose is extremely challenging and cannot be realized without careful planning for its requirements and the approaches to achieve the desired performance. There are several major obstacles to overcome before an ambulatory use of such a brain imager can become feasible. The first challenge is that PET is a nuclear medicine modality and one needs to deal with the issue of radiation dose and safety to the ambulatory healthy subjects and to the personnel. Therefore, we have to (1) develop a strategy to substantially lower the radiation doses. We hypothesize that doses as low as 1-10% of the standard injection dose when imaging with the whole body PET/CT scanners will become possible, without adversely impacting the task-specific accuracy of imaging procedures. The second challenge is (2) availability and delivery of the appropriate radiolabeled PET imaging agents that, with the substantially increased sensitivity of the device, will be able to tag neural targets that were too low to detect with current PET technology. The third key challenge is (3) the design of the mechanics/robotics allowing comfortable and safe freedom of movement. Clearly the most critical, lowering of the radiation dose, depends on many factors, starting with the definition of the task-specific performance requirements, and optimizations through simulation and development of optimized reconstruction algorithms, through selection of the optimized detector design, and of the type of the radioactive label (C-11, O-15 vs F18, etc.). This multi-parametrical analysis can be only achieved through careful analysis and simulation, accompanied by detector component prototyping, having in mind the follow-up grant application(s) to propose development of the full prototype. The first and critically important aim of our planning grant will be to articulate clearly the need for such an ambulatory, highly sensitive PET imager and prepare a list of its applications and accompanying task- specific performances. We prepared a preliminary list justifying the validity of the ambulatory PET concept, but we need to deepen it and associate operational requirements with particular imaging tasks. We anticipate that from our planning discussions we may envision more than one design of the imaging system, as one universal system may not suffice.